Monitoring query performance from Optimizer dashboard

Once the Query Optimizer is successfully activated for the Presto (C++) engine, it enables advanced query performance enhancements and optimization capabilities across multiple catalogs.

The Query optimizer manager allows collecting, storing, and synchronizing table and column statistics for specified tables. These statistics provide essential insights into the data, including row counts, the number of distinct values, minimum and maximum values for each column, and overall data distribution. Only Hive and Iceberg catalogs are supported for the collection of statistics.

The jobs described in this topic collect statistics and metadata and synchronize this information with the query optimizer, enabling more accurate and efficient query planning and execution. Only Hive and Iceberg catalogs are supported for the collection of statistics.

Viewing Presto (C++) engines

From the navigation menu, select Configurations and click the Query Optimizer Manager tile. The Query optimizer manager page opens.
Click Engines tab. It display the Presto (C++) engines, their status, and their associated catalogs.

Collecting statistical information from catalogs

You can do query analysis on catalogs (and tables) associated with the Presto (C++) engine.

From the navigation menu, select Configurations and click the Query Optimizer Manager tile. The Query optimizer manager page opens.
Click Catalogs tab.
The Statistics tab provides a view of associated catalogs, percentage of tables analyzed relative to the total number of tables within each catalog and the timestamp of the most recent analysis. You can filter the records based on the analysis. To do that, click the Filter icon. Select the following options and apply the filter condition:
- No analysis : Select this checkbox to filter catalogs that haven't been analyzed yet.
- Time since last analysis: Select this checkbox and provide the period (in days or months) to get the list of records processed during the specified period.
Click the Collect statistics button. The Collect statistics page opens with the schemas and tables associated with the Presto (C++) engines. Select the tables and click Create jobs to create statistics for tables.
You can also use the overflow menu next to each catalog on the Query Optimizer Manager page to initiate the creation of statistical data for the tables within the selected catalog.

View the status of the statistical jobs

From the navigation menu, select Configurations and click the Query Optimizer Manager tile. The Query optimizer manager page opens.
Click Catalogs tab.
The Statistics Jobs tab provides the list of statistical jobs that you created.

Active : Jobs initiated from the Collect Statistics page that is currently in a Running state is listed here. At any given time, only one job record will appear in the Active state. From the overflow menu, you can :
- View log : Select this to view the job log for the active record.
- Add to queue : Select this if you want to move the record to queue.
- Delete : Select this to delete the record.
The log information can include the following job status:

NOTRECEIVED: The system did not receive a call with a given task ID.

NOTRUN: An error prevented the scheduler from invoking the task’s procedure.

UNKNOWN: The task began execution, but the scheduler failed to record the outcome due to an unexpected condition.

Queued : If multiple jobs are initiated from the Collect Statistics page, only one will remain in the Active state at a time. All other jobs will be placed in a queue and listed here. From the overflow menu, you can select Remove from queue to remove the record from Queued state to Active state.

Completed : The jobs that have completed execution gets listed here. From the options available in the overflow menu, you can view the log, add the records to queue, or delete the records.